[vorbis-dev] Can compressed music sound better thanuncompressed?

[Steve Underwood]

Dirk Knop wrote:
> 
> Ahoy once more!
> 
> Steve wrote:
> >Many of these comparisons are performed using very mediocre amps and
> >speakers. Its quite possible that the compressed audio makes less
> >demands on the amp and speakers, as it has lost some detail. This meay
> >lead to an overall more pleasing, if less accurate, sound. It astonished
> >me to see the kind of crap reproduction equipment most people use to
> >here the glories of "digital quality".
> >
> >Another factor is that most people hear so little live music, that what
> >they hear through cheap audio equipment becomes the "correct" sound.
> 
> All your assumptions regarding these tests that I mean are wrong. They used
> all high end equipment and even reference design headphones etc.
> 
> The simplest conclusion is, that perceptual coding is flatening noise (for
> example: a rough distorted guitar may sound different after encoding
> because the distortion sounds "finer") etc. that may sound much better in a
> subjective way. It's clearly not a "good" (I mean precise) representation
> of the original sound anymore.
> 
> Believe me, that were high-end-equipped tests - that magazine i mentioned
> is the most respected one by "higher skilled" computer enthusiasts here in
> germany.

I'm not sure I follow you about all my assumptions being all wrong. You
are certainly disagreeing with my first point, but seem to be totally
agreeing with my second. People like what they are familiar with, and
that is generally poor quality sound. In fact, most people seem to be
pretty tone deaf, and can't really tell much difference between good and
bad audio, anyway. In a society that buys millions of 4inch speakers as
sub-woofers, can there really be a widespread appreciation of what good
sound quality audio actual is?

There are some aspects of music compressors which might well make the
job of the speakers easier. Mid to high end phase information is
generally discarded. If the final reproduced phase reduces the
percussive impact on the speaker cone of, say, plucking a string, it
will probably reduce the level of resonant oscillations in it. I'm not
sure if things really work out that way, though. Perhaps you could
engineer the decompressor specifically to help the mechanics of the
speakers.

Someone commented about the vinyl versus CD thing. Early CDs certainly
were a poor substitute for vinyl - the players had severe IM problems,
due to poor layout and poor DACs. Modern ones aren't perfect, and there
is possibly room to massage the signal to avoid the DAC's bad spots.
Again, whether current decompressors do that unintentionally I have no
idea. Interesting area for thought, though.

I was wondering recently how much a compressor could do to help the
speakers, by techniques not currently in wide use in music comrpessors.
It seems like you could bundle most of the very high frequency energy
into a single spectral line, and the ear wouldn't be able to tell the
difference. That would remove the need for the speaker to reproduce such
high frequency energy. Am I alone in such thoughts?

Regards,
Steve

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.